RPFbot: Unsupervised modular robot for precise feeding of rabbits

• A modular robot for precise feeding of rabbits has been developed. • The feed quantitative error is 1.73 % and the success rate in feeding is 100 %. • It is possible to provide rabbits in the same batch with different values of feed. • The RPFbot plays a significant role in large-scale breeding and rearing operations. In China’s efforts to implement the “Big Food Concept” for food security, the rapid development of the rabbit farming industry has become a key part of the livestock sector. Feeding systems are transitioning from traditional restrictive methods to precision feeding, with intelligent feeding equipment now a central trend in the industry. This study focuses on large-scale rabbit farming. The objective of the study was to develop the the Rabbit Precision Feeding Robot (RPFbot), which integrates mobile robot, precision feeding, robotic arm, and visual servoing technologies to overcome key challenges in rabbit feeding. RPFbot autonomously performs timed, differentiated, and precise feeding tasks, with feed quantification errors controlled to within 1.73 %. Field tests show excellent precision: the average error (unit: m) for the x-axis is 0.01295, the y-axis 0.00878, and the z-axis 0.00603. The Euclidean and Manhattan distance errors (unit: m) are 0.01896 and 0.02776, respectively, and the average joint error is 0.045337 rad, with a feeding success rate of 100 %. In terms of feeding efficiency, the average feeding times for the three rabbit cage feed boxes troughs corresponding to a single feeding point were 25 s, 23 s, and 21 s, respectively. RPFbot effectively prevents feed fragmentation, cross-contamination, and rabbit stress, while enabling differentiated feeding for rabbits within the same batch. This not only improves feeding efficiency but also supports optimized breeding, advancing the shift from automated to unmanned intelligent farms. The robot plays a crucial role in sustainable development and cost efficiency.